WO2012134440A1

WO2012134440A1 - Methods and devices for nucleic acid purification

Info

Publication number: WO2012134440A1
Application number: PCT/US2011/030232
Authority: WO
Inventors: Lee Hoang; Douglas T. Gjerde; Chris Suh
Original assignee: Phynexus, Inc.
Priority date: 2011-03-29
Filing date: 2011-03-29
Publication date: 2012-10-04
Also published as: GB2503856A; GB2503856B; US8921539B2; US20120252115A1; GB201318942D0

Abstract

The invention provides pipette tip columns for the purification of a DNA vector from unclarified cell lysate containing cell debris as well as methods for making and using such columns. The columns typically include a bed of extraction media positioned in the pipette tip column, above a bottom frit and with an optional top frit.

Description

METHODS AND DEVICES FOR NUCLEIC ACID PURIFICATION

FIELD OF THE INVENTION

This invention relates to methods and devices for sample preparation, such as separating

(i.e. , extracting or purifying) nucleic acids such as DNA and RNA, and more particularly plasmids. The device and method of this invention are particularly useful in DNA vector purification by pipette tip column. The device and method of this invention are particularly useful in purifying plasmids from un-clarified cell lysates and other samples containing particulates and cell debris.

BACKGROUND OF THE INVENTION

Commercially-available formats for nucleic acid purification include spin columns, magnetic beads in a tube or the use of vacuum to draw liquids through a column or plate. When nucleic acids are isolated from cells using these commercially-available formats, the cells are grown in a suitable medium, the culture is centrifuged to collect the cells and the growth medium is discarded. Next, the cells are lysed, e.g., with an alkali solution followed by the addition of a neutralizing solution. Traditionally, a second centrifugation step is performed after lysis to pellet the cell debris and the nucleic acids are purified from the supernatant.

When the spin column format is employed, several additional centrifugations are performed. Because these methods require a minimum of two centrifugation steps, they are time- consuming, laborious and difficult to automate. They require significant human intervention and cannot be performed in a walk-away fashion. Methods involving removal of the particulate from the cell lysate by filtration are not reliable. There exists a need for automated, high- throughput nucleic acid purification in a pipette tip column format. Furthermore, there exists a need for purifying plasmids from un-clarified cell lysates and other samples containing particulates and cell debris.

SUMMARY OF THE INVENTION

A highly automatable method for purifying nucleic acids in a pipette tip column format was developed. An advantage of the instant invention is that nucleic acids are purified after the lysis step without the need for cell debris removal. Nucleic acids are purified directly from an un-clarified lysate in an automated fashion. The method is particularly well suited for purification of plasmids. BRIEF DESCRIPTION OF THE FIGURES

Fig. 1 depicts an embodiment of the pipette tip column.

Fig. 2 depicts an embodiment of the vacuum block adapter with front and side views. Fig. 3 depicts an embodiment of the vacuum block adapter with top and bottom views. Fig. 4 depicts the layout of the deck of the Tecan Freedom Evo automated liquid handler.

DETAILED DESCRIPTION OF THE INVENTION

This invention relates to methods and devices for extracting nucleic acids, particularly plasmids from a sample solution. In U.S. Patent Application No. 10/620,155, now U.S. Patent 7,482,169, incorporated by reference herein in its entirety, methods and devices for performing low dead column extractions are described. In U.S. Patent Application No. 12/767,659, also incorporated by reference herein in its entirety, columns and methods for purification of DNA vectors are described.

Before describing the present invention in detail, it is to be understood that this invention is not limited to specific embodiments described herein. It is also to be understood that the terminology used herein for the purpose of describing particular embodiments is not intended to be limiting. As used in this specification and the appended claims, the singular forms "a", "an" and "the" include plural referents unless the context clearly dictates otherwise. Thus, for example, reference to polymer bearing a protected carbonyl would include a polymer bearing two or more protected carbonyls, and the like.

Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the invention pertains. Although any methods and materials similar or equivalent to those described herein can be used in the practice or testing of the present invention, specific examples of appropriate materials and methods are described herein.

It is a goal of the invention to develop an automated, high-throughput method for plasmid purification in a pipette tip column format. Commonly used commercially-available formats for plasmid purification include spin columns, vacuum plates and test tubes. However, there are currently no commercially-available, automated high-throughput methods performed in a pipette tip column.

In the invention described herein, plasmid DNA can be purified from any source. In some embodiments, the plasmids can be purified from biological sources such as cells. The cells can be eukaryotic or prokaryotic. In other embodiments, plasmids can be purified from a mixture of nucleic acids or from a gel. Plasmids are purified from multiple samples simultaneously in an automated manner for example, with a robotic workstation or electronic pipette. It is a goal of the invention is to reduce the number of manual processing steps used in methods for purifying plasmids. That is, it is desirable to perform high throughput separation of plasmids with minimal operator intervention. However, the invention also includes purification of plasmid DNA from a single sample.

To develop a robust method for purifying plasmid DNA from cells, experiments were performed in which plasmids were purified from E. coli cells. After the cells are grown, they are collected by centrifugation and the growth medium is discarded. The next step for purifying nucleic acids is cell lysis. Lysis can be carried out by a number of means including the use of chemicals i.e., detergents or by mechanical/physical means, such as sonication.

Currently, the predominant commercially-available formats for plasmid purification are spin columns, magnetic beads and vacuum plates. In these methods, cell debris is removed after the lysis step to obtain a "clarified lysate" from which nucleic acids are purified. Removal of cell debris is most often accomplished by centrifugation but is also done by filtration in some vacuum plate methods. However, because it is difficult to fully automate the steps of

centrifugation and filtration, it is time-consuming and laborious to purify nucleic acids from a large number of samples simultaneously.

Formats for plasmid preparation via vacuum include individual columns and multi-well plates. Even after producing a clarified lysate, these methods are not well suited for automation. Some protocols recommend turning off the vacuum while adding reagents, which requires operator involvement. Additionally, differences between samples can cause differential column pressures across the individual columns or wells within the plate so an operator is often needed to ensure the vacuum manifold seal is maintained or that the liquid sample flow occurs evenly through all the wells of the plate. Since spin columns require a series of centrifugation steps, they are not amenable to automation without special equipment. Magnetic beads are expensive and require repeated shake and aspiration steps, which makes their use difficult to automate. Magnetic beads or other bead suspension methods that do not first remove the cell debris are not reproducible and are difficult to automate.

An advantage of the instant invention is that plasmids can be purified in parallel, up to 96 samples at a time without operator involvement. With proper instrumentation, multiple plates of 96 samples can be processed simultaneously. Furthermore, the automated purification procedure begins with the cell pellet immediately after the lysis step. There is no need for cell debris removal. That is, nucleic acids are purified directly from an un-clarified lysate in an automated fashion. Automated methods are defined herein as methods not requiring human interaction. Excellent yield and concentration can be obtained using this method, e.g., a yield of 5 - 10 μg from a 1.4-mL E. coli culture harboring a low copy number plasmid.

Although it was desirable to eliminate the cell debris removal step and isolate nucleic acids directly from an un-clarified lysate, it was technically quite difficult to accomplish. Pipette tip columns provide a unique set of technical challenges not present in other formats such as spin columns or vacuum plates. To develop a robust, automated method for nucleic acid purification in a pipette tip column, it was necessary to overcome a number of technical obstacles. For example, when using a liquid handling robot, the pressure available to push liquids through the columns is very low compared to other methods such as centrifugation and vacuum.

In U.S. Patent Application No. 12/767,659, we described our first approach to solving these problems. Several steps were taken to achieve flow of the un-clarified lysate through the column. First, a number of different columns were produced by gradually increasing the frit pore size and increasing the packing particle size accordingly. Second, the column geometry was adjusted. The column diameter was increased, and the frit pore size was increased until the columns could tolerate some cell lysate particulates. Third, a pause step was incorporated into the pumping procedure. By pausing at the end of the each pumping stroke, the flow of the viscous lysate could catch up to the pump stroke. Fourth, entrance of particulates into the column was minimized by careful sampling of the unclarified lysate as the sample liquid passed though the column. The tip was placed low in the sample well so that the entire liquid volume could be processed.

Although the invention described in U.S. Patent Application No. 12/767,659 was an advance over other methods, the results obtained were still inconsistent. Sometimes, the columns would plug with the particulates contained in the un-clarified lysate. In some cultures, the particulates seemed to be greater in mass and all or most of the columns plugged. Even if the procedure worked without incident at times, the recovered, purified vector performed well for sequencing but sometimes couldn't be used effectively for transfection or transformation.

Another problem observed was that the A₂₆o was artificially high at times, particularly when the plasmid was present in a low or medium copy number. After plasmid purification, the concentration was measured by UV and also by a semi-quantitative measurement of the intensity of the plasmid band on an agarose gel. The comparison of these two methods suggested that something present in the sample might be co-purified with the plasmid, causing the A₂₆o to be artificially high.

In the instant application, these problems were solved making the method significantly more robust and reliable. The quality and purity of the product was improved making it useable for a greater variety of downstream applications. To address the problem of random column plugging and increase the reproducibility of the method, we examined and developed an entirely new sampling procedure. It was discovered that the amount and type of particulate in un-clarified lysate varied depending on a number of parameters including medium, strain, replicon, growth time and conditions. It turned out that the distribution of the cell debris present in the sample differed dramatically between samples. Sometimes the debris was distributed more or less throughout the sample, sometimes the majority of the debris floated, but in other instances a portion of the cell debris sank. This variability seemed to be one reason the method was not reproducible and that sometimes the columns plugged. Another reason seemed to be the amount of mass particulate varied tremendously from sample to sample. In some cases, the floating mass of particulate appeared to take up a large part, or even most of the sample.

Yet to recover the maximum amount of plasmid in the lysate, it was important to sample all of the liquid, regardless of where and how much particulate mass was in the sample.

Particulate masses present in the lysate contained liquid that appeared entrained and occluded. There did not appear to be active exchange of the occluded liquid with the other liquid in the sample.

Generally, in a suspension of particulates with liquid, the liquid can move freely throughout the sample. But when masses or globs of particulate accumulate in a sample in a stable form, free movement of the liquid within the mass is halted. The mass of particulate is almost like a large hydrated bead; there is no active transport of liquids but only diffusion. The masses looked globular and gel-like. It was speculated that plasmid contained in these globules would be unreachable unless the masses were broken up because active transport of liquid in and out of the mass would be limited. In U.S. Patent Application No. 12/767,659, passing these masses through the column broke up the masses and allowed capture of the plasmids. The only way to capture plasmid contained in the entire sample, including the sample within this occluded liquid, was to pass the entire sample through the column.

Improved sampling method

Several changes to the sampling method were studied to improve performance. First, the solutions were changed. In Patent Application No. 12/767,659, we used a lysis solution followed by a neutralization buffer comprised of a chaotropic salt, a salt and an acid. However, it was determined that it was more effective to use two solutions sequentially. The lysis solution was first followed by a solution for neutralization (acid and salt) and then a second solution containing the chaotropic reagent. When a solution containing salt and acid were added prior to the chaotropic salt solution, the Α₂₆ο in more accurately matched the plasmid concentration obtained by the slab gel band intensity. In addition, the amount of precipitate or cell debris generated seem to be more uniform. However, this did not solve the reproducibility and plugging issue. There were still large amounts of particulate masses in the sample that contained entrained liquid. In some cases, these masses floated or sometimes the masses precipitated out. Some particulate remained in suspension of the sample but depending on the cell growth conditions and time, the mass of cell debris appeared to make up about 20 - 50% of the sample.

Of course it is likely that the actual solids content in these masses was only a very small portion of the sample and that if the precipitates were centrifuged or filtered as in traditional methods, virtually all of the liquid could be recovered. Once spun down, the liquid is very easy to process using spin columns and plates. But having a substantial proportion of the sample entrained or occluded within the floating or sinking masses seemed to be the major issue. The liquid entrained within the mass of solid did not appear to be available for capture unless there was active transport of the liquid to the resin in the column. But unfortunately, it was found that the method taught in U.S. Patent Application No. 12/767,659 resulted in frequent plugging of the columns.

A second change made to the sampling procedure was that only a portion of the sample was aspirated and expelled. Instead of aspirating the entire un-clarified lysate, only a portion was sampled. Quite unexpectedly, it was determined that as little as 30 - 50% of the total volume could be repeatedly aspirated and expelled and the yield was not affected provided the number of cycles of liquid traveling through the column was increased. The term, "cycle" as used herein is defined as a single aspirate/expel step. Without being bound by theory, it is possible that the mass of particulate broke up and reformed with expulsion of the liquid back into the sample thus releasing or exchanging some of the entrained liquid. It did not seem possible that diffusion of the plasmid from the occluded liquid could occur because the distance to diffuse would be several millimeters and even as far as more than a centimeter in some cases.

Several side-by-side experiments were performed. Plasmid recovery was measured by A₂6o and by the plasmid band intensity on a slab gel. A side-by-side comparison of the method of the invention with commercial spin columns was performed. Also included in the side-by-side comparison was the old method of sampling where the entire liquid sample was passed through the column. The results from the slab gel band measurement showed that as the number of capture cycles was increased, the two pipette tip column methods gave comparable results while the spin columns gave slightly higher yield. The UV measurements at times gave comparable results for all three methods while at other times, the pipette tip column gave much higher results. Flow rates were adjusted to be slower until the all three methods gave UV results that agreed with the slab gel band intensity results. From this, it was surmised that at least part of the plasmid quality problem discovered earlier was due to the capture of sheared genomic DNA.

The experiments showed that sample volumes as low as 10% of the total volume in the well could be sampled and still get adequate sample recovery. As high as 90% of the volume could be sampled while still eliminating plugging of the column and get good recovery of the plasmid. Preferably 10 - 90 % of the sample volume can be sampled, more preferably 20 - 80% of the volume can be sampled, more preferably 30 - 70% of the volume can be sampled, more preferably 40 - 60 % of the volume can be sampled, most preferably 35 - 50% of the volume can be sampled. These results were unexpected and surprising in light of the fact that the particulates were often globular and appeared to have liquid sample entrained which had appeared to prevent capture of the plasmid within this liquid volume

In some embodiments, the sampling procedure was modified to include the addition of an aspirate and expel step prior to plasmid capture. Air is drawn slowly through the pipette tip columns attached to the robotic head. Then the columns are submerged in the sample and the air is slowly expelled through the columns into the un-clarified lysate. The step caused the bulk of the particulates to float which more effectively kept them farther away from the open lower end of the column during the subsequent aspirate/expel cycles used for plasmid capture.

Improved plasmid quality

The procedure described in U.S. Patent Application No. 12/767,659 yielded a suitable quantity of plasmid DNA that performed well in DNA sequencing, however it was discovered that the baculovirus transfection and bacterial transformation efficiency was unexpectedly low. Since the wash solution contains an organic solvent such as alcohol, it was suspected that residual wash solution might be present in the purified plasmid negatively impacted these processes. A refractometer was used to measure the alcohol content in the purified plasmid and it was approximately 20%. This result was surprising because it was thought that ethanol would prevent efficient elution of the plasmid from the column. In fact, the alcohol content of the elution solvent is low (zero) in order to get efficient elution.

In comparison, a mini -prep performed using a commercially-available spin column method produced a final alcohol content in the recovered plasmid in the 2 - 3 % range. So clearly, something about the columns or the method caused the results to be different. It was known that as the particle size of the resin used in the pipette tip columns was large. This was because the frit pore size of the columns had been increased to reduce plugging and therefore the particle size of the resin was also increased so that it did not fall out of the column. Without wishing to be bound by theory, it was known that the resin can contain pores to increase surface area and facilitate plasmid capture. Unfortunately, the resin appeared to retain much more solvent than the spin columns, possibly due to its higher porosity and greater surface area. This retained solvent may have contributed to the higher percentage of ethanol present in the eluted plasmid. Alternatively, the higher percentage of alcohol obtained from the pipette tip columns and method could have been due to some other unknown phenomena.

However, the automated purification of un-clarified lysates was desirable even with high alcohol content in the recovered plasmid. Several different remedies were tested to solve the problem of residual organic solvent in the purified plasmid. The first method evaluated was simply to lift the columns out of the wash solution and pass air back and forth through the column with the robotic pipette head. Even though the resin bed appeared to be equally wet at the beginning and end of the process, the amount of organic solvent in eluted plasmid decreased. While this method would likely work if the back and forth flow was performed with adequate number of cycles, it was not preferred because it added too much additional time to the method.

The next process tried was use of a 96-well aluminum heating block oven and a forced air oven. The ovens were set to 37 - 42°C. After final wash and expulsion of as much liquid as possible, the columns were placed in the ovens for 10 - 30 minutes. Again, the columns appeared wet after incubation in the ovens however, the ethanol concentration was reduced to as low as 5%. This was encouraging but the time required was still too long.

In a preferred embodiment to remove the organic solvent in an automated fashion, air was passed through the pipette tip columns. Air could be forced through the columns by positive pressure however, this would require an additional apparatus be designed and built. But it was also reasoned that vacuum could force air through the column by negative pressure the bottom end of the columns provided there were no leaks around the column body and the resistance to air flow due to negative pressure of the columns was consistent.

The vacuum method was investigated by depositing the pipette tip columns into a vacuum station on the robot deck and passing air through the columns using vacuum. An oil vacuum pump (0.5 horsepower) was used to pull a vacuum of 4 ft³/min through the columns. This use of vacuum is quite distinct from the traditional use of vacuum. Traditionally, vacuum is used to pull liquid solutions through plates or columns. After the solution passes through the plate or column, the vacuum is turned off because the task has been accomplished. In the case of the instant invention, the wash solution had already been passed through the columns and the vacuum is used simple to draw air through the columns.

It was necessary to build a custom 96-well vacuum block. To test the effectiveness of the vacuum block, it was necessary to build two additional air flow measurement apparatus. It was not possible to measure the air flow by seeing the liquid flow through the columns. The air flow had to be measured directly. The first apparatus was a cover for the vacuum block that was attached to and air gauge and used to measure air flow through the entire block. The air gauge (King Instrument Company, Part No. 75201102C17) was actually used in reverse. That is, air was pulled through the top of the gauge rather than being pushed through the bottom of the gauge as it was designed. Using this cover, a reading greater than 0.4 cfm was achieved with the pump and block being tested. Lifting the block from the vacuum manifold showed that there was a good seal between the vacuum block and its manifold base.

Several vacuum blocks were built before an adequate block design was found. The first block built had 96 positions on the top for the columns and an open architecture on the bottom of the block. The air flow through the block seemed adequate. However, it was not possible to get a tight seal when this block was tested with the cover. It seemed possible that while the total air flow may be adequate, the air flow across the individual columns could differ dramatically. A second custom apparatus was built to test the vacuum through individual columns seated in the vacuum block. In this case a bubble meter tube for measuring gas flow out of a packed bed gas chromatograph was modified to measure vacuum. A Wilmad LabGlass 10 mL gas flow bubble meter was adapted to measure air flow through the individual columns. As with the other gauge, the vacuum was applied to the top of the meter tube, leaving the tube fitting open that would normally have been the inlet from the gas chromatographic column. 96 columns were placed in the block and the air flow through each column was measured. Using this gauge, it was discovered that flow was not even between the columns. To solve this problem, the vacuum block was redesigned to have a gasket seal around each column.

The design of the column seal(s) proved to be difficult. The seal had to be tight enough to seal all of the columns routinely and adequately. But the column had to be easily placed into the apparatus and it must be possible to remove the columns from the block without the block being pulled up along with the columns. If the columns seals were too tight, attempting to remove the columns from the block could result in the block being lifted with the columns. So the seal could not be too tight. After several redesigns the block applied vacuum evenly through all the columns. Interestingly, it was not possible to determine whether air flowed through a particular column or not by visual inspection. Only the custom measurement tools could provide this information.

The next step involved testing the vacuum procedure for removal of the organic solvent present in the wash solution. Liquid containing various amounts of ethanol was cycled through the columns. The columns were placed in the vacuum block and vacuum was applied for varying amounts of time. The columns were eluted with water, the eluant collected and the refractive index was measured for organic solvent concentration. Depending on the vacuum applied and the air flow through the individual columns, the "drying time" could be lowered to 2-5 minutes. Based on these experiments, a vacuum duration was determined for which the eluant contained less than 6%, less than 5%, less than 4%, less than 3% or less than 2% organic solvent. Less than 5% solvent was preferred and less than 3% was most preferred.

After the solvent removal step, the columns still appeared wet by visual inspection. To maintain the highest possible throughput, it was desirable to find the shortest possible vacuum duration that resulted in purified plasmid having acceptably low alcohol content. Although the solvent drying step is an additional step to the process, if a very strong vacuum is used, the columns can be dried more quickly without sacrificing throughput.

However in other embodiments with longer drying conditions, it is possible to dry the columns completely prior to elution of the purified plasmid. However, good results were obtained when only the solvent was removed and the columns were not dried completely after the final wash and prior to elution.

Unfortunately, implementation of the solvent removal step negatively impacted the reproducibility of the elution step. For example, when 80 μL· bed silica columns were subjected to vacuum to remove the organic solvent present in the wash, 130 μL· of water was used for the elution step and only 80-90 μL· of liquid was collected. This result indicated that a significant portion of the water was trapped in the dead volume of the pipette tip column.

To mitigate this problem, the elution step was modified. A blow-out step was added at the end of the procedure to maximize the elution volume. To perform the blow-out step, air is taken up by the robotic pipetting head after the solvent removal step and prior to engagement of the columns for elution. This added air is then expelled after expulsion of the purified plasmid to get as much liquid as possible out of the column. Yield was also increased by incubating the elution solution on the column prior to expulsion. As an example, the elution solvent can be aspirated and incubated on the column for 5 minutes prior to expulsion.

Scale-up of recovered plasmid

Another significant point of novelty of the instant invention is in the area of scale-up. Plasmid purification protocols are typically called "mini-prep", "midi-prep" or "maxi-prep" based on their scale. Although these plasmid purification protocols are well known in the art, an automated system for performing 96 midi-preps at a time has never been described. In the instant application, scale -up to a "midi-prep" culture was achieved.

In a mini-prep embodiment, up to 20 μg of the plasmid could be recovered. In another embodiment of the instant invention, a midi-prep scale-up of the purification process was achieved so that 96 samples were processed simultaneously with a plasmid DNA yield of 20 to 100 μg. An automated procedure for processing samples in 96-well format and producing 20 to 100 μg of plasmid DNA from has never before been accomplished.

Development of this parallel, automated midi-prep procedure required additional technical obstacles to be overcome. For the purpose of this invention, a mini-prep automated approach is defined as a method in which the amount of plasmid or nucleic acid recovered is in the 1-20 μg range. A midi-prep automation approach is defined as a method where the amount of plasmid or nucleic acid recovered is in the 20-100 μg range.

One of the biggest technical obstacles in developing the automated midi-prep columns and method was scaling. Because of the constraints imposed by automation, it was possible to simply scale up all the columns and solution volumes.

For example, a commercially-available mini-prep spin column has bed dimensions of 7.0 mm diameter and 2.05 mm height giving a bed volume of 79 mm³. When the bed material is scaled to midi-prep size, the bed dimensions increase to 13.9 mm diameter and 11.2 mm height giving a bed volume of 1700 mm³. This is more than a 20 fold increase in bed size. The preferred bed volume of the mini-prep scale in the instant invention is 20 - 120 μL· bed with the 60 - 80 μL· bed size most preferred. Certainly the bed size can be decreased if a lower recovery of plasmid is desired. But if a larger amount of plasmid is required, it is not possible to increase the 60 - 80 μL· preferred 20 fold to 1.2 mL - 1.6 mL bed size when using a 96-well format. Pipette tips used with robotic liquid handling stations are limited to volumes of 0.2 mL, 1.0 mL or 1.2 mL. These pipette tips cannot hold enough medium for a 20-fold scale-up. Even if the columns were designed to be used with deep-well plates the body size could only be 2 mL at most. There would not be enough room above the resin bed to process the sample, wash solutions and elution solvent.

According to commercially- available columns, the volumes of the solutions were scaled

15 to 20 fold when scaling from mini-prep to midi-prep. Clearly, this was an added difficultly when scaling an automated method performed in a 96-well format. 2 mL deep well plates are the most common size for the 96-well format, although 4 mL plates are available. Multiple wells could be used to contain the solutions but this quickly becomes impractical as the solution volume approaches 16 mL and higher. At the outset, it appeared to be impossible to scale the automated method to obtain 20-100 μg of purified plasmid.

The first problem tackled was the size of the column bed. It was known that the bed size could not be too large because of limited chamber space above the bed. In one embodiment, a 300 μL· resin bed in a 1 mL pipette tip was tested. This bed height was 3.75 times higher than the mini-prep columns (80 μL· resin bed) described herein. In another embodiment, the column bed size was 400 μL·.

Experiments were performed in which enough cell lysate was passed through the column to load the columns to capacity without consideration of the volumes of resuspension buffer, lysis buffer, precipitation buffer, wash and elution solutions needed. Surprisingly and unexpectedly, it was discovered that the resin did have enough capacity to recover up 100 μg of plasmid. Without wishing to be bound by theory, the yield may have been due to the porous nature of the packing material. Nevertheless the results were unexpected. The column bed size for midi-prep recovery of 20 - 100 μg nucleic acid recovery ranged from 100 - 800 μL·, 200 - 500 μΐ. or 300 - 400 μί.

Next the volume constraints of the sample, resuspension buffer, the lysis buffer and the precipitation buffer were examined. After the cells are resuspended, too much lysis solution would be needed to effectively lyse the large number of cells would be too high without modification. It is true that the sample could be captured from a number of wells, perhaps up to 4 or more, but at some point the larger volume become difficult to work with.

In one embodiment, midi-prep procedure employs a 4 mL resuspension buffer, 4 mL lysis buffer, and 6 mL of precipitation buffer, making the total volume 14 mL. This could be acceptable for a 24 well format, but would require 8 wells of a 96-well deep-well block. So while this embodiment would be possible to automate, it is not preferred.

In order to solve this issue, several smaller lysis volumes were tested to reduce the total volume we need to process the midi sample:

1. 300 μL· re-suspension, 300 μL· lysis, 410 μL· precipitation: total=1010 μL·

2. 500 μL· re-suspension, 500 μL· lysis, 700 μL· precipitation: total=1700 μL·

3. 1 mL re-suspension, 1 mL lysis, 1.4 mL precipitation; total=3.4 mL

4. 2 mL re-suspension, 2 mL lysis, 2.8 mL precipitation: total=6.8 mL

The 500 μL· re-suspension volume method is preferred. In one embodiment, 30 mL of overnight grown culture and performing is processed by the midi-prep. In another embodiment the starting culture volume is between 5-15 ml which produced approximately 50 μg of purified plasmid. With low density cultures, it is desirable to start by processing 30 mL of culture to get recoveries of greater than 50 μg of plasmid.

The smaller bed size facilitated the use of smaller wash volumes and elution volumes. This is the first known automated method that produces plasmid DNA at the midi-prep scale of up to 100 μg of plasmid DNA. The 96-well automated method can be performed with clarified or un-clarified lysates. Definitions

In describing and claiming the present invention, the following terminology will be used in accordance with the definitions set out below.

The term "bed volume" as used herein is defined as the volume of medium or solid phase within the column. The term "interstitial volume" of the bed refers to the volume of the bed of extraction media that is accessible to solvent, e.g., aqueous sample solutions, wash solutions and desorption solvents. This includes the space between the beads as well as any volume taken up by the beads by the pores. The interstitial volume of the bed represents the minimum volume of liquid required to saturate the column bed.

The term "dead volume" as used herein with respect to a column is defined as the interstitial volume of the extraction bed, tubes, membrane or frits, and passageways in a column. Some preferred embodiments of the invention involve the use of low dead volume columns, as described in more detail in U.S. Patent 7,482,169.

The term "elution volume" as used herein is defined as the volume of desorption or elution liquid into which the analytes are desorbed and collected. The terms "desorption solvent," "elution liquid" and the like are used interchangeably herein.

The term "frit" as used herein is defined as porous material for holding the medium in the column. In preferred embodiments of the invention, the frit is a thin, low pore volume, large pore screen.

The term "pipette tip column" as used herein is defined as any column containing a solid phase that can engage a pipette or syringe or liquid handler, either directly or indirectly. The term, "pipette tip column" is not limited to columns manufactured in pipette tips. Rather the column can have any shape or geometry as long as it is capable of engaging a pipette, syringe pump or liquid handling robot. Pipette tip columns can be located in a rack or be incorporated into a plate.

The term "lysis" or "lysed" is a process by which cell are treated to break the cell walls and release the nucleic acid. In some methods lysis procedure involves treatment with an alkali solution followed by the addition of a neutralizing solution. The neutralization solution may contain an acid. It may also contain a chaotropic agent and/or other components. In some methods, the lysis procedure involves treatment with a surfactant. In some methods, the lysis procedure is mechanical or physical.

The term, "plasmid" is defined as an extra-chromosomal, self-replicating nucleic acid molecule. A plasmid can be a single or double stranded and can be comprised of DNA or RNA. Cosmids, BACs and YACs are considered to be within the purview of the plasmid definition. The columns

In the subject invention, a bed of medium is contained in a column, wherein the bed is held in place with a bottom frit. In some embodiments, the columns are additionally comprised of a top frit. Non-limiting examples of suitable columns, particularly low dead volume columns are presented in U.S. Patent 7,482,169. It is to be understood that the subject invention is not limited to the use of low dead volume columns. The columns may be configured into plates or racks or used individually.

Typically, the column is comprised of a column body having an open upper end, an open lower end, and an open channel between the upper and lower ends of the column body; a bottom frit extending across the open lower end and a bed of medium positioned inside the open channel above the bottom frit.

Figure 1 depicts an embodiment of pipette tip column construction. Disposable pipette tip 160 is cut approximately ¼ inch from the lower end and frit 174 is welded to the lower end of the tip body. A silica resin 184 was then transferred into the tip. In certain embodiments, upper frit 198 is placed above the resin, e.g., using a friction fit. The lower end is removed from a second pipette tip 190 and the remaining upper end is inserted into pipette tip 160 and held in place by a friction fit. Pipette tip 190 is specific for the liquid handling system that will be used to process the columns. In some embodiments, pipette tip 190 is additionally comprised of barrier 182. In other embodiments, barrier 182 is absent. Barrier 182 is particularly useful when frit 198 is absent because it serves to confine resin 184 within the pipette tip column during shipping.

Media

The chemistry employed in the present invention is typically based on normal phase or ion-exchange. Ion-pairing may also be used for nucleic acid purification. Because the invention is directed to the purification and/or concentration of nucleic acids, extraction surfaces capable of adsorbing such molecules are particularly relevant. In general, these chemistries, methods of their use, appropriate solvents, etc. are well known in the art.

The media used in the column is preferably a form of water-insoluble particle (e.g., a porous or non-porous bead) that has an affinity for the nucleic acid of interest. Silica beads are suitable for the columns of the invention. Chromosorb P is large and works well. Silicon quartz also works well. Other suitable materials include celite, diatomaceous earth, silica gel, silica gel, (Davisil, Impaq, Biotage), metal oxides and mixed metal oxides, glass, alumina, zeolites, titanium dioxide, zirconium dioxide.

The bed volume of the medium used in the columns of the invention is typically in the range of 10 μL· and 500 μL·, 10 μL· and 300 μL·, 20 μL· and 100 μL·, or between about 15 μL· and 80 μL·. For midiprep scale the bed volume of the medium used in the columns of the invention is typically in the range of 100 μL· and 800 μL·, 100 μL· and 300 μL·, 200 μL· and 300 μL·, or between about 200 μΐ. and 400 μί. The average particle diameters of beads of the invention are typically in the range of about 20 μιη to several hundred micrometers, e.g., diameters in ranges having lower limits of 20 μιη, 30 μιη, 40 μιη, 50 μιη, 60 μιη, 70 μιη, 80 μιη, 90 μιη, 100 μιη, 150 μιη, 200 μιη, 300 μm, or 500 μιη, and upper limits of 10 μιη, 20 μιη, 30 μm, 40 μm, 50 μιη, 60 μm, 70 μιη, 80 μm, 90 μm, 100 μιη, 150 μm, 200 μιη, 300 μm, 500 μm, 750μιη, or 1 mm.

The space between resin particles can also be important. This space increases with looser packing of the column. Preferred beds are not tightly packed.

Frits

One or more frits is used to contain the bed of medium in a column. Frits can take a variety of forms, and can be constructed from a variety of materials. The frits of the invention are porous, since it is necessary for fluid to be able to pass through the frit. The frit should have sufficient structural strength so that frit integrity can contain the extraction media in the column. It is desirable that the frit have little or no affinity for chemicals with which it will come into contact during the extraction process, particularly the analyte of interest. Frits of various pores sizes and pore densities may be used provided the free flow of liquid is possible. Frits of pore size large enough to prevent plugging from cell debris are of particular interest.

In one embodiment, a single frit (e.g., a lower, or bottom, frit) extends across the open channel of the column body. Preferably, the bottom frit is attached at or near the open lower end of the column, e.g. , extending across the open lower end. Normally, a bed of medium is positioned inside the open channel in contact with the bottom frit.

In certain embodiments, a top frit may be employed. For example, in some embodiments, a second frit extends across the open channel between the bottom frit and the open upper end of the column body. In this embodiment, the top frit, bottom frit and column body (i.e., the inner surface of the channel) define a media chamber wherein a bed of medium is positioned. The frits should be securely attached to the column body and extend across the column body to completely occlude the channel, thereby substantially confining the bed of medium inside the media chamber.

In other embodiments, the top frit is positioned well above the medium, e.g., 15 - 20 mm or more above the medium. The position of the top frit can be proximal to open upper end of the pipette tip column. That is, the top frit can be closer to the open upper end than to the bed medium. In these embodiments, the bed is not packed and the medium can occupy well under 50% of the volume of the extraction media chamber. In these embodiments, the top frit can be significantly thicker than the bottom frit and liquids may not flow through the top frit.

The position of the top frit over the bed may just touch the top of the resin bed or be positioned substantially above the resin bed. When the frit is above the resin bed, the resin bed may move or expand with aspiration of liquids including the sample containing the particulates. The bed may move down against the bottom frit with expulsion of the liquid.

In some preferred embodiments of the invention, the bottom frit is located at the open lower end of the column body. This configuration is not required, i.e., in some embodiments, the bottom frit is located at some distance up the column body from the open lower end. Some frits of the invention have a large pore size frit.

The performance of the column is typically enhanced by the use of frits having pore or mesh openings sufficiently large to allow cell debris or other particulates to flow through the frit without clogging or plugging under low pressures applied by a pipette or liquid handler. Of course, the pore or mesh openings of course should not be so large that they are unable to adequately contain the extraction media in the chamber. Frits of the invention preferably have pore openings or mesh openings of a size in the range of about 5 - 500 μιη, more preferably 10 - 200 μιη, and still more preferably 100 - 150 μιη, e.g., about 120 μιη.

In some cases, it is necessary to consider the relationship between the frit pore size and the particle diameter. Specifically, it is possible to increase the frit pore size when the particle diameter is increased. For example, a frit pore size of 100 μιη was used successfully with a range of different resins.

Some embodiments of the invention employ a thin frit, preferably less than 1000 μιη in thickness (e.g., in the range of 20 - 1000 μιη, 40 - 350 μιη, or 50 - 350 μιη), more preferably less than 200 μιη in thickness (e.g., in the range of 20 - 200 μιη, 40 - 200 μιη, or 50 - 200 μιη), more preferably less than 100 μιη in thickness (e.g., in the range of 20 - 100 μιη, 40 - 100 μιη, or 50 - 100 μιη). However, thicker frits, up to several mm, 5 and even 10 mm, thick may be used if the pore size of the frit can be increased dramatically. Increasing the frit thickness can only be done if the pore size of the is increased.

Some preferred embodiments of the invention employ a membrane screen as the frit. The use of membrane screens as described herein typically provide this low resistance to flow and hence better flow rates, reduced back pressure and minimal distortion of the medium. The membrane can be a woven or non-woven mesh of fibers that may be a mesh weave, a random orientated mat of fibers i.e. a "polymer paper," a spun bonded mesh, an etched or "pore drilled" paper or membrane such as nuclear track etched membrane or an electrolytic mesh (see, e.g. , 5,556,598). The membrane may be, e.g., polymer, glass, or metal provided the membrane is low dead volume, allows movement of the sample and various processing liquids through the column bed, may be attached to the column body, is strong enough to withstand the bed packing process, is strong enough to hold the column bed of beads, and does not interfere with the extraction process i.e. does not adsorb or denature the sample molecules.

The frit can be attached to the column body by any means which results in a stable attachment. For example, the screen can be bonded to the column body through welding or gluing. The column body can be welded to the frit by melting the body into the frit, or melting the frit into the body, or both. Alternatively, a frit can be attached by a friction fit or by means of an annular pip, as described in U.S. Patent 5,833,927.

The frits of the invention can be made from any material that has the required physical properties as described herein. Examples of suitable materials include polymer, sintered polymer, fiber, nylon, polyester, polyamide, polycarbonate, cellulose, polyethylene,

nitrocellulose, cellulose acetate, polyvinylidine difluoride, polytetrafluoroethylene (PTFE), polypropylene, polysulfone, PEEK, PVC, vinyl polymer, metal, ceramic and glass.

In certain embodiments of the invention, a wad of fibrous material is included in the column, which extends across the open channel below the open upper end of the column body, wherein the wad of fibrous material and open channel define a media chamber, wherein the medium is positioned within the media chamber. This wad of fiber can be a porous material of glass, polymer, metal, or other material having large pores. In some embodiments, the wad of fibrous material is used in lieu of an upper frit.

Solvents

Disruption of bacterial cell membranes is typically accomplished using an alkaline solution containing a detergent. Any detergent that effectively disrupts the cell membrane can be used for this purpose.

In certain embodiments of the invention, chaotropic agents can be added to the sample prior to plasmid capture. Examples of chaotropic reagents include sodium iodide, sodium perchlorate, guanidine thiocyanate (GuSCN), urea, guanidine hydrochloride (GuHCl), potassium iodide, sodium perchlorate, potassium chloride, lithium chloride, sodium chloride, urea or mixtures of such substances.

Examples of suitable solvents for use with the invention are shown in Tables A and B. TABLE A

Normal Phase

Normal Phase Reverse Phase

Chaotropic

Extraction Ion-Pair Extraction

Extraction

Typical solvent

Low to medium High to medium High to medium polarity range

Typical sample Hexane, toluene, chaotropic buffers H₂0, buffers, ion- loading solvent CH₂CI₂ alcohol pairing reagent

H₂0/CH₃OH, ion-

Typical desorption

pairing reagent solvent Ethyl acetate,

H₂0/CH₃CN, ion- acetone, CH₃CN

pairing reagent

(Acetone,

(Methanol, acetonitrile,

H₂0/buffer chloroform, acidic isopropanol,

methanol, basic methanol, water,

methanol, buffers)

tetrahydrofuran, acetonitrile, acetone, ethyl acetate)

Sample elution Least polar sample Most polar sample Most polar sample selectivity components first components first components first

Solvent change Increase solvent Decrease Decrease solvent required to desorb polarity chaotropic buffer polarity

TABLE B

Methods for using the columns

The invention provides a pipettor (such as a multi-channel pipettor) suitable for acting as the pump in methods such as those described herein. In some embodiments, the pipettor comprises an electrically driven actuator. The electrically driven actuator can be controlled by a microprocessor, e.g., a programmable microprocessor. In various embodiments the

microprocessor can be either internal or external to the pipettor body. The back pressure of a column will depend on the average bead size, bead size distribution, average bed length, average cross sectional area of the bed, back pressure due to the frit and viscosity of flow rate of the liquid passing through the bed. For a column described in this application, the backpressure at 2 mL/min flow rate ranged from 0.05 to 2 psi.

In certain embodiments of the method, the sample, wash and or desorption solvents are aspirated and discharged from the column more than once, i.e., a plurality of in/out cycles are employed to pass the solvent back and forth through the bed more than once. In some embodiments, throughput is maximized by performing some steps with bidirectional flow and other steps by vacuum, pressure or gravity flow. For example, the capture step can be performed using bidirectional flow and the wash and elution steps can be performed using vacuum or gravity flow. In these embodiments, the robotic pipetting head can be used more efficiently for dispensing liquids, allowing a greater number of columns to be processed in parallel.

In preferred embodiments of the invention, a plurality of columns is run in a parallel fashion, e.g., multiplexed. Multiplexing can be accomplished, for example, by arranging the columns in parallel so that fluid can be passed through them concurrently. When a pump is used to manipulate fluids through the column, each column in the multiplex array can have its own pump, e.g., syringe pumps activated by a common actuator. Alternatively, columns can be connected to a common pump, a common vacuum device, or the like.

In certain embodiments the pipettor is a multi-channel pipettor. However, in preferred embodiments, a robotic system such as those commercially available from Zymark, Hamilton, Beckman, Tecan, Packard, Matrix, Phy Nexus, Agilent and others are used for plasmid purification. Those robots having a 96-channel pipetting head are particularly preferred.

The invention also provides software for implementing the methods of the invention. For example, the software can be programmed to control manipulation of solutions and addressing of columns into sample vials, collection vials, for spotting or introduction into some device for further processing.

The invention also includes kits comprising one or more reagents and/or articles for use in a process relating to solid-phase extraction, e.g., buffers, standards, solutions, columns, sample containers, etc.

The open lower end of the pipette tip column can be positioned relatively close to the corresponding well bottom, e.g., within a range having a lower limit of about 0.05 mm, 0.1 mm, 0.2 mm, 0.3 mm, 0.4 m, 0.5 mm, 1 mm, 2 mm, 3 mm, 4 mm, or 5 mm from the bottom of the well, and an upper limit of 0.3 mm, 0.4 m, 0.5 mm, 1 mm, 2 mm, 3 mm, 4 mm, 5 mm, 6 mm, 7 mm, 8 mm, 8 mm or 10 mm of the well bottom. For example, in some embodiments the open lower end of a pipette tip column is positioned with between 0.05 and 2 mm from a well bottom, or between 0.1 and 1 mm from a well bottom. The term "well bottom" does not necessarily refer to the absolute bottom of a well, but to the point where the tip makes contact with the well when the tip is lowered to its full extent into the well, i.e., a point where the tip can seal with the well surface. For example, in some microwell plate formats the wells taper down to an inverted conical shape, so a typical tip column will not be able to make contact with the absolute bottom of the well. Purification of plasmid DNA from E. coli

Nucleic acids and particularly plasmids can be purified from any source including eukaryotic or prokaryotic cells, tissues, body fluids (blood, serum, plasma, saliva, urine, feces), tissue culture, bacteria, viruses. The purification procedure can be used with low, medium or high copy number plasmids. The instant invention can also be used to isolate nucleic acids from a gel.

When purifying plasmid DNA from E. coli, the first step is cell growth. A person of skill in the art can select the appropriate growth conditions depending on the cell type, number of samples, desired yield, etc. For example, bacterial cells can be grown at 37°C in a 96-well deep- well block with shaking at 300 rpm and harvested in the late logarithmic stage of growth. The deep-well block can be selected according to the desired culture volume. For example, a 4-ml deep well block can be used if a larger cell culture is required. Generally, a rich medium is used such as Terrific Broth, 2xYT or Agencourt Ale (Beckman Coulter) containing the appropriate antibiotic. After the cells are grown, they are centrifuged and the growth medium is discarded.

The next step involves resuspension of the cells e.g., in a buffer. From this point, the remainder of the procedure can be fully automated with the use of a liquid handling system. In those embodiments in which the procedure is automated, buffer is added and the cell suspension is repeatedly aspirated and expelled from a pipette tip until the cells are completely resuspended. Alternatively, the resuspension step may be performed manually by vortexing until the pellet is fully resuspended.

After resuspension, the next step is cell lysis. Lysis can be accomplished by a number of means including physical or chemical action. Non-limiting examples of lysis methods include mechanical, such as ultrasonic waves, mortar and pestle, osmotic shock, chemical e.g. by means of detergents and/or chaotropic agents and/or organic solvents (e.g. phenol, chloroform, ether), heat and alkali. Lysis via chemical means can be performed on a liquid handling system by addition of a lysis solution to the resuspended cells.

A precipitation buffer is added to the lysed cell suspension to precipitate the genomic DNA prior to capture. In preferred embodiments, the precipitation buffer is comprised of chaotropic salts. After lysis, the nucleic acid is captured using a pipette tip column. In existing methods, a centrifugation step is usually performed following cells lysis to pellet cell debris. However, an advantage of the instant invention is that this centrifugation step can be bypassed, making the method considerably more automated than other methods.

The column can be equilibrated with water or buffer prior to the capture step.

Equilibration can be performed by a single aspiration and expulsion of water or buffer from the column. After the pipette tip columns are equilibrated, the plasmid is captured on the equilibrated column by repeated aspiration and expulsion.

After capture, the plasmids bound to the column are usually washed to remove non- specifically bound materials. One or more wash steps can be performed. When more than one wash is performed, the same wash solution can be used for multiple washes or different wash solutions can be used. In certain embodiments, the wash solution contains an organic solvent, e.g., alcohol.

Wash steps can be performed with back and forth flow or unidirectional flow using gravity or vacuum. The advantage of performing the wash steps by unidirectional flow is that higher throughput can be achieved. That is, when plasmid purification is performed on a liquid handling robot, throughput can be increased by utilizing the liquid handling head simply for dispensing wash solution to multiple plates. When the wash is performed by back-and-forth flow, the liquid handling head can only process one plate at a time.

After the wash step, air is passed through the columns to remove any organic solvent remaining from the wash step. This can be accomplished by depositing the pipette tip columns onto a vacuum block and drawing air through the columns with a vacuum. A vacuum block adaptor was custom built for this process and is described in more detail below.

In certain embodiments, air is passed through the columns long enough to remove the organic solvent present in the wash solution, but not long enough to dry the columns completely. When the residual organic solvent is measured, it is in the range of less than 10%, less than 9%, less than 8%, less than 7%, less than 6%, less than 5%, less than 4%, less than 3%, less than 2% or less than 1%.

In other embodiments, the columns can be dried completely. Complete drying is not preferred, because it requires more time. However, if the pump is sufficiently strong, the columns can be dried completely without sacrificing significant time.

In still other embodiments, air is passed through the columns with positive pressure. Alternatively, it is possible to dry or remove the ethanol or other organic solvent after elution by methods such as speed-vac, air drying, heating or applying a gas stream to the wells containing the eluted sample.

The elution of plasmid from the column can be accomplished with back and forth flow or one direction flow. Generally elution volumes are in the range of about 1 - 5 times the bed volume. When back-and-forth flow is used, air can be aspirated into the pipette tip column prior to aspirating the elution buffer. This air can be used after expulsion of the plasmid to ensure complete expulsion of all the liquid in the column Generally, the elution buffer is aqueous and has a pH between 6 and 10. In some embodiments, the column is incubated with the elution buffer for a period of time. In these embodiments, the column and elution buffer are incubated for at least 1 minute, at least 2 minutes, at least 3 minutes, at least 4 minutes, at least 5 minutes, at least 10 minutes or at least 15 minutes. In other embodiments, the incubation step is omitted.

After the incubation step, the purified plasmid is expelled from the pipette tip column. To ensure the maximum volume of purified plasmid is recovered, a blow-out step can be performed by expelling the air aspirated as described above.

The concentration of plasmid DNA purified by this method is generally at least 50 ng/uL, at least 75 ng/uL, at least 100 ng/uL or greater and an A260 280 ratio of 1.8 - 2.0. Most importantly, the plasmid DNA purified by these methods is high quality and can be used for any downstream application including sequencing, transfection and transformation.

The vacuum block adapter

The vacuum adapter block shown in Figures 2 and 3 can have several functions. It is a plate that contains the pipette tip column and is fitted over a vacuum manifold. In some embodiments, vacuum applied to the manifold can be controlled by software to apply vacuum to the columns at the appropriate time for the appropriate duration.

Figure 2 depicts embodiments of the side and front views of the vacuum adapter block and Figure 3 shows the top and bottom of the block. In this embodiment, the block contains positions for 96 columns. In other embodiments, the block may contain positions for any number of columns including 8, 12, 24 or 48 columns.

Figure 2A depicts an embodiment of the side view of the adapter block, the tops of eight columns 10 are inserted into top block 20. Top block 20 is separated from bottom block 40 by sealing gasket 30. The gasket serves to seal around each individual column when they are inserted into the block so that the vacuum is applied through the columns and not around the sides of the column bodies. The bottom of bottom block 40 contains plastic lip 50. In this embodiment, the lip conforms to SBS standardized format for 96-well plates so that the base of the block can be inserted into the vacuum manifold or any deck position of a robotic liquid handler.

Figure 2B depicts the front view of the vacuum adapter block pictured in Figure 2A. It is identical to the side view shown in view A except that the row of twelve columns 10 can be seen.

Figure 2C is a cut-away view of the vacuum block adapter front view. Pipette tip columns 60 are exposed to show that when inserted into the block they extend almost to the bottom. In this embodiment, the end of the column does reach the bottom of the vacuum block. In other embodiments, the lower ends of the columns will be even with the block. In other embodiments, the ends of the columns will extend out past the base of the vacuum block.

Opening 100 allows the vacuum to be applied at the bottom of the block and allows liquid and air passage through the columns sealed by gasket 30. Cross section of top block 70 is separated from cross section of bottom block 80 by gasket 30. In certain embodiments, the column shape is frustoconical and the holes at the interface of top block 70 and bottom block 80 have a smaller diameter than those on the upper surface of top block 70. The plasmid is captured by column packing material 90, washed and eluted from the sample.

Figure 3 depicts and embodiment of the top view of the vacuum adapter block and Figure 3B shows the bottom view of the vacuum block. Lip 50 lies at the bottom of block near bottom surface 130. Pipette tip columns are inserted into through holes 110 from top surface 120.

All publications and patent applications mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication or patent application was specifically and individually indicated to be incorporated by reference.

Having now generally described the invention, the same will be more readily understood through reference to the following examples, which are provided by way of illustration, and are not intended to be limiting of the present invention, unless so specified. EXAMPLES

The following preparations and examples are given to enable those skilled in the art to more clearly understand and practice the present invention. They should not be construed as limiting the scope of the invention, but merely as being illustrative and representative thereof. Example 1

Evaluation of an 80 bed volume pipette tip column containing a resin for purification of plasmid from eukaryotic cells

In this example, the performance of 80 μL· bed volume pipette tip columns is evaluated. The pipette tip column was constructed from a 200 μL· pipette tip (Tecan) and is packed with a silica-based particle resin. These columns, buffer conditions and column processing procedures are tested for the recovery of plasmid DNA from yeast. The yield and quality are assessed by UV spectrometry and agarose gel electrophoresis.

Samples are prepared by growing a single yeast colony in 25 mL medium supplemented with the appropriate carbon source to propagate the DNA vector. The liquid culture is incubated at 30⁵C with shaking until the culture becomes turbid. The culture is divided into equal aliquots and subjected to centrifugation at 5,000 x g for 15 minutes to pellet the yeast. The supernatant is discarded and the pellets are lysed by mortar and pestle, using liquid nitrogen and resuspended in buffer.

To purify the plasmid DNA from the lysed yeast cells, the pipette tip columns are processed by the ME semi-automated purification system (PhyNexus, Inc., San Jose, CA). The columns are equilibrated with 200 μΕ 7M guanidinium-HCl by performing one cycle of back- and- forth flow at 500 μΕ/min and a 20 second pause at the end of the aspirate and dispense steps.

The yeast lysate is subjected to pipette tip column processing for capture of the plasmid DNA by using at least 24 back-and-forth cycles at a flow rate of 250 μΕ/min with 20 second pauses after the end of each aspirate and dispense step.

Following plasmid capture on the pipette tip column, the columns are washed with 200 μΕ wash 1 buffer consisting of 10 mM Tris-HCl pH 6.6, 5M guanidinium-HCl and 30% ethanol. This is followed by a second wash in wash 2 buffer consisting of 10 mM Tris-HCl pH 7.5 and 80% ethanol. Both wash procedures are carried out using one cycle of back-and-forth flow at a flow rate of 500 μΕ/min with 20 second pauses at the end of each aspirate and expel step. A blow out step is incorporated to remove all residual wash buffer from the resin bed.

DNA plasmid is released from the column with 300 μΕ elution buffer consisting of water. The procedure to release the DNA is 8 back-and-forth cycles at a flow rate of 250 μΕ/min with 20 second pauses after the end of each aspirate and dispense step.

Example 2

Purification of plasmid DNA from E. coli

Columns and methods for purifying plasmid DNA from E. coli lysate were developed for 96 samples at a time. The columns used in this example were 80 μΕ bed columns fitted with 100 μιη pore size screen bottom frits. The method was designed to operate on a Tecan EVO, Biomek FX or other robotic liquid handler. The solutions used are listed in Table 1

E. coli cells were grown to late logarithmic phase, harvested by centrifugation and then resuspended in buffer. The plasmid purification procedure developed was as follows.

1. Add 250 μΕ of Lysis buffer to resuspended cells using gentle pipette mixing for 3 minutes. 2. Add 350 μΕ of Neutralization buffer to lysed culture using gentle pipette mixing for 3 minutes.

3. Attach plasmid DNA pipette tip columns to 96 channel head.

4. Equilibrate the pipette tip columns by cycling through the equilibration buffer (2.8 min).

- Use 2 cycles at 0.5 mL/min flow rate.

5. Capture the plasmid DNA (50 min). - Use 24 cycles at 0.25 mL/min flow rate.

6. Wash (Washl buffer, 500μί) the captured plasmid DNA (2.8 min).

- Use 2 cycles at 0.5μί/ιηίη flow rate.

7. Wash (Wash2 buffer, 500μΙ.) the captured plasmid DNA (2.8 min).

- Use 2 cycles at 0.5μΙ7ιηίη flow rate.

8. Wash (Wash2 buffer, 500μΙ.) the captured plasmid DNA (2.8 min).

- Use 2 cycles at 0.5 mL/min flow rate.

9. Blowout remaining wash buffer.

10. Elute the captured plasmid DNA (33 min).

- Use 16 cycles at 0.25 mL/min flow rate.

Total time -95 min

The yield was greater than 5 μg per well. The purity was examined with slab gel electrophoresis and UV absorption with A260 A280 ratio between 1.8 and 2.0. The DNA was used successfully for transfection and had no RNA contamination.

Example 3

Comparison of pipette tip columns and spin columns

The pipette tip columns used in this example contained 80 μL· of Chromosorb P resin (Sigma Aldrich) and were fitted with 105 μιη pore size screen bottom frits. A side by side comparison with commercial spin columns was made using buffers listed in Table 1. E. coli was grown overnight in 1.4 mL growth in a 96 deep well plate. The results of three representative samples are shown in Table 2.

TABLE 1 Buffers

Buffer Name Content

Resuspension 50 mM Tris-HCl pH 8.0, 10 mM EDTA, 100 ug/mL RNase A

buffer

Lysis buffer 200 mM NaOH, 1% SDS

Neutralization 4.2M guanidine hydrochloride

buffer 0.9M Potassium acetate Ph 4.8

Equilibration 7M Gu-HCL pH 5.5

buffer

Washl buffer 5M guanidine hydrochloride

30% Ethanol, 10 mM TRIS-HCl pH 6.6

Wash2 buffer 10 mM TRIS-HCl pH 7.5, 80% Ethanol

Elution buffer Water

Table 2 Comparison of pipette tip columns and spin columns

Representative results from purification of plasmid using three identical commercial spin columns (Spin CI, Spin C2 and Spin C3) and two types of pipette tip columns. El, E2 and E3 refer to the recovery from three sequential elution aliquots, elutions 1 through 3.

Example 5

Mini-prep of E. coli plasmid DNA from 96 samples at a time

Single colonies were inoculated into 1.4 ml rich medium (containing the appropriate antibiotic) in a 2-ml deep- well block and incubated at 37°C and 300 rpm for 16 hours. The deep- well block was centrifuged and the medium was discarded. The plate was then transferred to a Tecan Freedom Evo with the deck set up as follows and shown in Fig. 4.

Positions 1 through 3 contain boxes of 200-μΕ pipette tips. Position 4 has a box of 96 pipette tip columns. In this example the pipette tip columns are constructed with a bottom frit only (pore size 105 μιη) and filled with 80 μΕ of silica resin. Position 5 holds a 96-well plate filled with 250 μΕ Precipitation Buffer in each well. Positions 6 and 7 contain plates holding lysis and resuspension buffers, respectively. Positions 8, 11 and 12 contain buffers for wash 1, wash 2 and wash 3, respectively. Each of these is a deep-well block holding 500 μΕ of buffer. A deep-well block holding 300 μΕ Equilibration buffer is placed in position 9. The deep-well plate holding the cell pellets is placed at position 10. There is a UV-readable plate at position 13 to receive the purified plasmid DNA. Stations 14 and 15 can be used for drawing air through the pipette tip columns with vacuum and a UV plate reader resides at position 16.

The plate was processed as follows.

1. Resuspend cells. Transfer 150 ul resuspension buffer to cell pellet. 130 ul, 8 - 16 cycles, 10 ml/min.

2. Lyse cells. Add 150 uL of Lysis buffer to resuspended cells. 8 cycles of 180 μΕ at 10 ml/min with 2 sec pause.

3. Add 210 μΕ precipitation buffer. 8 cycles of 180 μΕ at 10 ml/min with 2 sec pause.

4. Attach pipette tip columns to 96 channel head. Equilibrate the pipette tip columns. 2 cycles of 180 μΕ, 0.5 ml/min with 5 sec pause.

5. Capture

a. Aspirate 200 μΕ air at 0.25 ml/min with 2 sec pause

b. Submerge pipette tip column in unclarified lysate and expel 200 μΕ air at 0.25 ml/min with 2 sec pause. Particulates should float.

c. Capture. 180 μΕ of unclarified lysate, 14 cycles at 0.25 ml/min with 20 sec pause.

6. Wash 1. 180 μΕ of wash buffer 1, 2 cycles at 0.5 ml/min with 10 sec pause.

7. Wash 2. 180 μΕ of wash buffer 2, 2 cycles at 0.5 ml/min with 10 sec pause.

8. Wash 3. 180 μΕ of wash buffer 2, 2 cycles at 0.5 ml/min with 10 sec pause.

9. Vacuum dry. Deposit tips to vacuum station and vacuum air through the tips for 5 min.

10. Elution. a. Aspirate 70 μL· of air.

b. Engage tips and aspirate 130 μΕ of elution buffer at 0.25 ml/min.

c. Incubate 5 min.

d. Expel 130 μΕ of purified plasmid at 0.25 ml/min.

TABLE 3 Solutions

Example 6

Procedure for Midi-prep of E. coli plasmid DNA from 96 samples at a time

The buffers used in this example are listed in Table 3.

1) In 10-30 mL of LB media, inoculate a single colony.

2) Grow overnight. 37°C, 16 hours at 300 rpm.

3) Centrifuge for 25 minutes at 3000 rpm.

4) Discard the supernatant.

5) Resuspend pellet with 1 mL resuspension buffer containing 0.4 mg/mL RNase A.

6) Add 1 mL of Lysis buffer. Mix thoroughly.

7) Add 1.4 mL of Precipitation buffer. Mix thoroughly.

8) Attach PhyTip columns to the ME/MEA and equilibrate in 500 uL of equilibration buffer.

The columns contain a 300 bed in a 1 mL pipette tip (2 cycles at 0.5 ml/min)

9) Intake 1 mL air into the column at a flow rate of 0.5 ml/min.

10) Move the pipette tip column to the bottom of the precipitated sample.

11) Expel 1 mL of air at 10 ml/min.

12) Capture plasmid by performing 10 - 15 cycles (0.25 ml/min or 0.5 ml/min). 13) Three wash steps. Move the pipette tip columns into a deep well block containing 1 mL of wash buffer. 4 cycles (0.5-5 ml/min).

14) Air dry. Use vacuum pump. 5 - 10 minutes.

15) Move the pipette tip columns into the deep well block containing 1.2 ml of elution buffer. 16) Intake 1.2 mL, wait 5 min and expel 1.2 mL.

Example 7

Midi-prep of E. coli plasmid DNA using a combination of back and forth flow and gravity flow.

In this example, the midi-prep is performed as described in the preceding example except the wash and elution steps are done using gravity flow. For the wash step, only Wash buffer 2 is used. The column is washed with 1 ml of buffer and the wash step is repeated 10 - 15 times. For the elution step, 1.2 mL of elution buffer is used. Example 8

Example of extraction of DNA from agarose gel

The nucleic acids in this example are not limited to plasmid DNA. This procedure can be used to isolate nucleic acids of any type or size distribution that can be visualized on a gel.

Agarose gel electrophoresis is the most common method for size separation and visualization of double stranded DNA. Agarose gels are used to separate DNA based on length of the DNA.

Shorter DNA migrates farther through the gel compared to a long DNA. In practice, agarose gels are used to purify PCR products away from free primers, dNTPs, DNA polymerase and buffer components. The PCR product will migrate as a discreet band. Restriction digests of plamids, for example, also result in discreet bands that can be purified by agarose gel. Discreet bands correspond to DNA of the same length. To utilize this separation as a pre-purification tool, the band corresponding to the DNA length of interest is excised from the gel using a scalpel or razor blade. The band is weighed and is placed into a microfuge tube. Three volumes of gel extraction buffer (50 mM MOPS pH 7.0, 1M NaCl, 15% (v/v) isopropanol) is added to the excised gel using the conversion 1 mg = 1 μί. The tube is incubated at 50 ⁵C for 10 minutes. The tube is vortexed every 2 to 3 minutes during this incubation. One volume of isopropanol is added to the tube.

A plasmid DNA pipette tip purification column is used to capture the DNA. The column is processed by the PhyNexus MEA personal purification instrument. The MEA engages the pipette tip column and equilibrates it with 2 cycles of back-and-forth flow in water using a flow rate of 0.5 mL/min and 20 second pauses at the end of each aspirate and dispense step. Next, the column captures the extracted DNA. This is accomplished using 4-20 cycles of back-and- forth flow at a flow rate of 0.25 mL/min and 20 second pauses at the end of each aspirate and dispense step. The columns are subject to a wash in 0.5 mL Wash Buffer (80% ethanol, 10 mM Tris-HCl pH 7.5). The wash is repeated in fresh buffer an additional two times. After washing, the pipette tip columns are transferred to a vacuum block and subject to 5 minutes of vacuuming to dry the columns to remove residual Wash Buffer components. The MEA next engages the pipette tip columns and aspirates 130 \lL of water and incubates for 5 minutes. This is dispensed to release the plasmid DNA and a second elution is performed if necessary.

While the invention has been described in connection with specific embodiments thereof, it will be understood that it is capable of further modifications and this application is intended to cover and variations, uses, or adaptations of the invention that follow, in general, the principles of the invention, including such departures from the present disclosure as come within known or customary practice within the art to which the invention pertains and as may be applied to the essential features hereinbefore set forth. Moreover, the fact that certain aspects of the invention are pointed out as preferred embodiments is not intended to in any way limit the invention to such preferred embodiments.

Claims

CLAIMS What is claimed is:

1. An automated method for purifying plasmids with a pipette tip column, the method

comprising:

i) providing a plurality of un-clarified liquid samples comprised of plasmid DNA, liquid and particulates;

ii) providing a plurality of pipette tip columns, wherein the pipette tip columns contain a solid phase having an affinity for plasmid DNA;

iii) passing the samples through the pipette tip columns;

iv) optionally, repeating step (iii);

v) passing a wash solution through the pipette tip columns, wherein the wash solution is comprised of an organic solvent;

vi) optionally, repeating step (v);

vii) passing air through the pipette tip columns; and

viii) eluting the plasmids by passing an aqueous solution through the pipette tip columns, wherein the organic solvent content of the eluted plasmid is 5% or less.

2. The method of claim 1, wherein one or more of the sample solution, the wash solution and the elution solution is passed through the pipette tip columns by gravity flow.

3. The method of claim 1, wherein one or more of the sample solution, the wash solution and the elution solution is passed through the pipette tip columns using vacuum.

4. The method of claim 1, wherein one or more of the sample solution, the wash solution and the elution solution is passed through the columns by aspirating and expelling the solution through the open lower end of the pipette tip columns.

5. An automated method for capturing plasmid DNA from an unclarified lysate with a pipette tip column, the method comprising:

i) providing an unclarified lysate having a sample volume, wherein the unclarified lysate is comprised of plasmid DNA, liquid and particulates;

ii) providing a pipette tip column;

iii) submerging the open lower end of the pipette tip column in the unclarified lysate; iv) aspirating and expelling a portion of the unclarified lysate through the open lower end of the pipette tip column, wherein the portion of the unclarified lysate is not more than 80% of the sample volume; and

v) optionally, repeating step (iv).

6. The method of claim 5, wherein the portion of the unclarified lysate is not more than 60% of the sample volume.

7. The method of claim 6, wherein the method is performed on at least 2 and at most 96 pipette tip columns simultaneously.

8. The method of claim 7, wherein between steps (ii) and (iii), air is aspirated into the pipette tip columns and between steps (iii) and (iv) air is expelled from the pipette tip columns.

9. The method of claim 1, wherein step (iii) is performed by repeatedly aspirating and expelling a portion of the sample through the open lower end of the pipette tip columns.